Workshop: Semantic Foundations for Molecular Biology
Schemata, Controlled Vocabularies and Ontologies
The Sixth International Conference on Intelligent Systems for Molecular Biology (ISMB-98)
Sponsored by the International Society for Computational Biology
Tutorial: Saturday, June 27, 1998 (evening)
Workshop: Sunday, June 28, 1998 (all day)
Molecular biology has a communication problem. Many researchers and databases use (at least partially) idiosyncratic terms and concepts for representing biological information. Often, terms and definitions differ between groups, with different groups not infrequently using identical terms with different meanings. The concept "gene", for example, is used with different semantics by the major international genomic databases. Ontologies are one way to provide a semantic repository of systematically ordered relevant concepts in molecular biology. Such repositories can be used, for example, to bridge the different notions in various databases by explicitly specifying the meaning of and relation between fundamental concepts.
An ontology has been variously defined as:
"A specification of a conceptualization." (T. Gruber)
"A conceptualization is a set of models of L (a given logical language)which describes the admittable (intended) interpretations of its non logical symbols (the vocabulary). An ontology is a (possibly incomplete)axiomatization of a conceptualization." (N. Guarino)
"An ontology is a particular knowledge base, describing facts assumed to be always true by a community of users: (1) by virtue of the agreed-upon meaning of the vocabulary used (analytical knowledge) and (2) whose truth does not descend from the meaning of the vocabulary used (nonanalytical, common knowledge)." (N. Guarino)
"An ontology is a (possibly incomplete) axiomatization of the intended models of a logical language." (N. Guarino)
"Concise and unambiguous description of principal, relevant entities of an application domain and their potential relations to each other." (S. Schulze-Kremer)
"An ontology is a specification of a vocabulary of non-logical symbols and includes: types of entities, attributes and properties, relations and functions, and constraints." (R. Fikes)
"A reusable domain theory." (N.N.)
"The portion of a knowledge base that does not change during problem solving." (N.N.)
"A formal specification of the terms and relationships among those terms for some domain of knowledge." (N.N.)
Ontologies are now recogized as essential to information integration and knowledge-based systems in a wide variety of disciplines. This is because ontologies permit knowledge workers to define and share domain-specific vocabularies. Having developed a formal specification for a domain ontology, it is possible for database and software developers to agree on its use. Biology, including molecular biology, genetics, and biochemistry have historically played host to many nomenclature wars and a vast quantity of rather loosely defined terminology. In a rapidly developing scientific discipline, a certain amount of terminological fluidity cannot be avoided, but the level of conceptual casualness has resulted in large and essential databases that are both inconsistent internally and unmanagable as parts of a larger information infrastructure for research.
Part I (Saturday evening) - Tutorial
This tutorial will introduce participants to the background and recent developments in ontology research, ontology specification languages, ontology building methods and tools, and some examples of the use of ontologies in information systems and for semantic integration of diverse information sources in molecular biology. This introduction will help to resolve and assimilate the widely varying terminology in this field and to provide a common basis for the following day workshop.
Part II (Sunday) - Workshop
This one-day workshop will focus on the following aspects of schemata, controlled vocabularies and ontologies for molecular biology and bioinformatics:
formal representations of conceptual structures
practical requirements for molecular biology
curation, maintenance and accessibility
re-use and sharing
compatibility with legacy systems
development of standards
tools and languages
The workshop will consist of interactive case studies of conceptual models from existing systems. One goal of the workshop is to assess the current success and "division of labor" among researchers, and evaluate the possibilities for interoperability.
We would also be particularly interested in hearing about content areas of biology that have been modeled, the strengths and weaknesses of those models and modeling techniques used, and to what extent the modeling principles are applicable to other domains.
The precise format will depend on the number and background of interested participants. Participants will be encouraged to develop, present, discuss, merge, and begin the implementation of plans for collaborations and consortia to jointly develop interoperable ontologies for molecular biology. Position papers proposing standards to facilitate ontology interoperation are also encouraged. A more detailed schedule will appear here as it becomes available.
Russ B. Altman Stanford University
W. David Benton, SmithKline Beecham
Peter D. Karp, Pangea Systems Inc.
Steffen Schulze-Kremer, Max-Planck-Institute for Molecular Genetics
Attendees of the tutorial will have to register with ISMB-98. Registration fee for the tutorial is $100 ($40 for students. There is a reduced rate if an additional tutorial is booked). Payment to ISMB can be made by check, money order or credit card.
Attendees of the workshop will have to register with the Max-Planck-Institute for Molecular Genetics, Berlin. Registration fee for the one-day workshop is $50 ($40 for students or if you are also registered for the tutorial "Ontologies for Molecular Biology"). Payment should be made in cash at the workshop. Sorry, no credit cards.
To indicate interest in the workshop (and be added to the workshop mailing list) contact Steffen Schulze-Kremer with a one paragraph statement of your interest/experience with some or all of these issues at the addresses indicated below.
Dr. Steffen Schulze-KremerMax-Planck-Institute for Molecular GeneticsDepartment LehrachIhnestrasse 73D-14195 BerlinGermany
Tel + 49 (0)30 / 8413-1426Fax + 49 (0)30 / 8413-1384
March, 31st 1998